The Agent Skills Directory

[SAFE]: The skill instructions and associated Python scripts do not contain any malicious patterns or security vulnerabilities.
[COMMAND_EXECUTION]: The scripts agent.py and process.py are designed for local facilitation of exercises, including scenario generation and scoring, without invoking arbitrary shell commands or external processes.
[DATA_EXFILTRATION]: There are no network-related functions or commands that access sensitive files; all operations are confined to the exercise data.
[EXTERNAL_DOWNLOADS]: All external links point to well-known and trusted official domains for cybersecurity standards (CISA, FEMA, NIST) for educational purposes.
[PROMPT_INJECTION]: The skill instructions provide legitimate guidance for training and testing organizational readiness without attempting to bypass AI safety filters or override agent behavior.
[REMOTE_CODE_EXECUTION]: The skill lacks any functionality for downloading and executing remote code or performing dynamic evaluation of untrusted input.

performing-ransomware-tabletop-exercise